Design Of Flexible Exoskeleton Structure Of Ankle Joint And Research On Assistance Under Variable Speed Conditions

At present,the aging trend of the world’s population is becoming increasingly serious,and the number of disabled people is increasing year by year.It is of great significance to design a low-cost and comfortable lower limb assistance exoskeleton suitable for promotion to the general public.Most single-joint robots only need one or two power sources,with relatively simple structure,light weight and relatively low cost,which is more suitable for general promotion.When the human body walks normally,the ankle joint is the main force application joint,so this paper mainly studies the assistance of the ankle joint.In this paper,a flexible ankle exoskeleton is designed from the perspective of human simulation.The specific research contents are as follows:Based on the analysis of the current situation of ankle joint development,the overall mechanism design scheme of ankle joint,including the design scheme of driving mechanism,transmission mechanism and executive mechanism,is proposed with the design goal of structural simplification,cost saving and strong wearability.On this basis,the finite element analysis of the main stressed parts of the ankle joint is carried out in combination with the design objective to verify the rationality of the scheme.Put forward the basic concept of upper control,study the feasibility of dividing the human gait information through the plantar pressure sensor,study the feasibility of filtering the noise of the plantar pressure sensor by the Kalman filter,study the feasibility of obtaining the assist torque according to the gait information combined with the torque curve formula fitted by the subsection function method,and study the percentage of the gait cycle as the input signal of the formula,Based on the feasibility of the optimization scheme of the middle foot pressure sensor as the gait periodic adjustment signal,the concept of infinite subdivision is proposed,and the feasibility of combining probabilistic neural network(PNN)and radial basis function neural network(RBF)with gait information to detect the output torque is studied.The lower control system puts forward the concept of variable PID control system based on the ankle joint torque curve and human walking law.On this basis,it studies the principle of redundant force generation,the acquisition method of ankle joint expected speed when human walking,and the feasibility of variable PID force loading system control scheme based on speed compensation.Determine the experimental scheme and build an experimental test platform for wearable ankle flexible exoskeleton.In the upper control,verify the feasibility of subsection function scheme and its optimization scheme and neural network system,and in the lower control,verify the feasibility of variable PID control strategy and variable PID control system based on speed compensation.